github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-07-03 03:38:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

community[patch]: Using the right encoding to parse the web page in RecursiveUrlLoader (#20632)

Browse Source 
As shown in #13749 , `RecursiveUrlLoader` has encoding issue. This PR is
to solve this.

---------

Co-authored-by: Bagatur <22008038+baskaryan@users.noreply.github.com>
Co-authored-by: Bagatur <baskaryan@gmail.com>
This commit is contained in:
fubuki8087 2024-05-01 02:41:36 +08:00 committed by GitHub
parent b0b1a67771
commit f1c3687aa5
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 15 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
libs/community/langchain_community/document_loaders/recursive_url_loader.py
View File

@ -98,6 +98,8 @@ class RecursiveUrlLoader(BaseLoader):
continue_on_failure: bool = True, continue_on_failure: bool = True,
*, *,
base_url: Optional[str] = None, base_url: Optional[str] = None,
autoset_encoding: bool = True,
encoding: Optional[str] = None,
) -> None: ) -> None:
"""Initialize with URL to crawl and any subdirectories to exclude. """Initialize with URL to crawl and any subdirectories to exclude.
@ -137,6 +139,11 @@ class RecursiveUrlLoader(BaseLoader):
continue_on_failure: If True, continue if getting or parsing a link raises continue_on_failure: If True, continue if getting or parsing a link raises
an exception. Otherwise, raise the exception. an exception. Otherwise, raise the exception.
base_url: The base url to check for outside links against. base_url: The base url to check for outside links against.
autoset_encoding: Whether to automatically set the encoding of the response.
If True, the encoding of the response will be set to the apparent
encoding, unless the `encoding` argument has already been explicitly set.
encoding: The encoding of the response. If manually set, the encoding will be
set to given value, regardless of the `autoset_encoding` argument.
""" # noqa: E501 """ # noqa: E501
self.url = url self.url = url
@ -148,6 +155,8 @@ class RecursiveUrlLoader(BaseLoader):
if metadata_extractor is not None if metadata_extractor is not None
else _metadata_extractor else _metadata_extractor
) )
self.autoset_encoding = autoset_encoding
self.encoding = encoding
self.metadata_extractor = _wrap_metadata_extractor(metadata_extractor) self.metadata_extractor = _wrap_metadata_extractor(metadata_extractor)
self.exclude_dirs = exclude_dirs if exclude_dirs is not None else () self.exclude_dirs = exclude_dirs if exclude_dirs is not None else ()
@ -184,6 +193,12 @@ class RecursiveUrlLoader(BaseLoader):
visited.add(url) visited.add(url)
try: try:
response = requests.get(url, timeout=self.timeout, headers=self.headers) response = requests.get(url, timeout=self.timeout, headers=self.headers)
if self.encoding is not None:
response.encoding = self.encoding
elif self.autoset_encoding:
response.encoding = response.apparent_encoding
if self.check_response_status and 400 <= response.status_code <= 599: if self.check_response_status and 400 <= response.status_code <= 599:
raise ValueError(f"Received HTTP status {response.status_code}") raise ValueError(f"Received HTTP status {response.status_code}")
except Exception as e: except Exception as e: